ACSELW
	ACSET
	ACSETM
	ACSETMW
	ACSETW
	ACSINC
	ACSINCW
	ACSINV
	ACSINVW
	ACSNEG
	ACSNEGW
	ADC
	ADCPS1
	ADCPS2
	ADCPS3
	ADMB
	ADRPS
	ADSB
	ADWORD
	AEON
	AEONW
	AEOR
	AEORW
	AERET
	AEXTR
	AEXTRW
	AFABSD
	AFABSS
	AFADDD
	AFADDS
	AFCCMPD
	AFCCMPED
	AFCCMPES
	AFCCMPS
	AFCMPD
	AFCMPED
	AFCMPES
	AFCMPS

	DMB $0                                     // bf3003d5
	DRPS                                       // e003bfd6
	DSB  $1                                    // 9f3103d5
	EONW R21<<29, R6, R9                       // c974354a
	EON R14>>46, R4, R9                        // 89b86eca
	EOR $-2287828610704211969, R27, R22        // 76e343d2
	EORW R12->27, R10, R19                     // 536d8c4a
	EOR R2<<59, R30, R17                       // d1ef02ca

TEXT ·p256MovCond(SB),NOSPLIT,$0
	MOVD	res+0(FP), res_ptr
	MOVD	a+8(FP), a_ptr
	MOVD	b+16(FP), b_ptr
	MOVD	cond+24(FP), R3

	CMP	$0, R3
	// Two remarks:
	// 1) Will want to revisit NEON, when support is better
	// 2) CSEL might not be constant time on all ARM processors
	LDP	0*16(a_ptr), (R4, R5)
	LDP	1*16(a_ptr), (R6, R7)
	LDP	2*16(a_ptr), (R8, R9)
	LDP	0*16(b_ptr), (R16, R17)

	case lex.LSH:
		op = 0
	case lex.RSH:
		op = 1
	case lex.ARR:
		op = 2
	case lex.ROT:
		// following instructions on ARM64 support rotate right
		// AND, ANDS, TST, BIC, BICS, EON, EOR, ORR, MVN, ORN
		op = 3
	}
	tok := p.next()
	str := tok.String()
	var count int16
	switch tok.ScanToken {
	case scanner.Ident:
		if p.arch.Family == sys.ARM64 {